def process(tag_name, tc, save_flag=False, try_times=2000):
""" 学習とその検証を繰り返して、結果をファイルに保存する
"""
_dir = create_dir([tag_name])
with open(tag_name + "/av_verify_report" + tag_name + ".csv", "w") as fw:
for i in range(try_times):
print("--try count: {0}/{1}--".format(i, try_times))
# 教師データを作成
verify_data, teacher_dates, x, y = None, None, None, None
if save_flag:
_save_name_teacher = tag_name + "/av_teaching_data" + tag_name + "_{0:05d}".format(i) + ".csv" # tc.save_teacher(ファイル名)で教師と検証データを保存するときに使う
_save_name_verify = tag_name + "/av_verify_data" + tag_name + "_{0:05d}".format(i) + ".csv"
verify_data, teacher_dates, x, y = tc.save_teacher(save_name_teacher=_save_name_teacher, save_name_verify=_save_name_verify) # 毎回、呼び出すたびにデータセットの内容が変わる
else:
verify_data, teacher_dates, x, y = tc.create_dataset() # 毎回、呼び出すたびにデータセットの内容が変わる
features_dict = tc.get_all_features()
training_data = (x, y)
# 学習と保存
clf, score = learning.learn(training_data)
mc.save(clf, tag_name)
fw.write("{0},{1}\n".format(i, score))
# 過学習の確認（現状では役に立っていない）
dates = sorted(verify_data.keys()) # 学習に使っていない日付
if len(dates) == 0:
continue
result = predict.predict2(clf, dates, features_dict) # 学習に使っていないデータで検証
def main():
X, Y = load_data()
classifier, iters, error, (precision, recall, f1) = learning.learn(X, Y)
print("Best C: {0}".format(iters))
print("Test set error: {0}".format(error))
print("Precision: {0}".format(precision))
print("Recall: {0}".format(recall))
print("F1 score: {0}".format(f1))
def main():
X, Y = load_data()
classifier, iters, error, (precision, recall, f1) = learning.learn(X, Y)
print("Best C: {0}".format(iters))
print("Test set error: {0}".format(error))
print("Precision: {0}".format(precision))
print("Recall: {0}".format(recall))
print("F1 score: {0}".format(f1))
def create_clf(clfque):
if os.path.isfile('clf3.pckl'):
clf = pickle.load(open('clf3.pckl', 'rb'))
else:
clf = lern.learn()
clf_file = open('clf3.pckl',
'wb') # saving the clf variable into memory
pickle.dump(clf, clf_file)
clf_file.close()
clfque.put(clf)
def solveOne(setN, setName, ansName):
startTime = time()
maxRating, nUsers, nItems, trainX, trainY, testX, testY = loadData(setName, ansName)
print("Set #{0}: read {1} users, {2} items, {3} train set ratings and {4} test set ratings".format(setN, nUsers,
nItems,
len(trainX),
len(testX)))
bestL, predictedRs = learn(maxRating, nUsers, nItems, trainX, trainY, testX)
print("Set #{0}: RMSE = {1}, best λ = {2}".format(setN, errors.rmse(predictedRs, testY), bestL))
endTime = time()
print("Set #{0}: Leaning done in {1} seconds".format(setN, endTime - startTime))
def main():
startTime = time.time()
X, Y = load_data()
classifier, iters, error, (precision, recall, f1) = learning.learn(X, Y)
print("Best L: {0}".format(iters))
print("Test set error: {0}".format(error))
print("Precision: {0}".format(precision))
print("Recall: {0}".format(recall))
print("F1 score: {0}".format(f1))
endTime = time.time()
print("Done working in {0} seconds".format(endTime - startTime))
def solveOne(setN, setName, ansName):
startTime = time()
maxRating, nUsers, nItems, trainX, trainY, testX, testY = loadData(
setName, ansName)
print(
"Set #{0}: read {1} users, {2} items, {3} train set ratings and {4} test set ratings"
.format(setN, nUsers, nItems, len(trainX), len(testX)))
bestL, predictedRs = learn(maxRating, nUsers, nItems, trainX, trainY,
testX)
print("Set #{0}: RMSE = {1}, best λ = {2}".format(
setN, errors.rmse(predictedRs, testY), bestL))
endTime = time()
print("Set #{0}: Leaning done in {1} seconds".format(
setN, endTime - startTime))
def process(tag_name, tc, save_flag=False, try_times=2000):
""" 学習とその検証を繰り返して、結果をファイルに保存する
"""
_dir = create_dir([tag_name])
with open(tag_name + "/av_verify_report" + tag_name + ".csv", "w") as fw:
for i in range(try_times):
print("--try count: {0}/{1}--".format(i, try_times))
# 教師データを作成
verify_data, teacher_dates, x, y = None, None, None, None
if save_flag:
_save_name_teacher = tag_name + "/av_teaching_data" + tag_name + "_{0:05d}".format(
i) + ".csv" # tc.save_teacher(ファイル名)で教師と検証データを保存するときに使う
_save_name_verify = tag_name + "/av_verify_data" + tag_name + "_{0:05d}".format(
i) + ".csv"
verify_data, teacher_dates, x, y = tc.save_teacher(
save_name_teacher=_save_name_teacher,
save_name_verify=_save_name_verify
) # 毎回、呼び出すたびにデータセットの内容が変わる
else:
verify_data, teacher_dates, x, y = tc.create_dataset(
) # 毎回、呼び出すたびにデータセットの内容が変わる
features_dict = tc.get_all_features()
training_data = (x, y)
# 学習と保存
clf, score = learning.learn(training_data)
mc.save(clf, tag_name)
fw.write("{0},{1}\n".format(i, score))
# 過学習の確認（現状では役に立っていない）
dates = sorted(verify_data.keys()) # 学習に使っていない日付
if len(dates) == 0:
continue
result = predict.predict2(clf, dates,
features_dict) # 学習に使っていないデータで検証
def main():
####### Estimate Distance ########
vision.img_capture()
distance = vision.img_distance()
print("Distance estimated ", distance)
if distance == 0:
print("Distance estimation Failed! ")
exit(0)
####### Get KNN Model ########
knn_model = learning.learn()
####### Predict hit Success/Failure ########
d = distance
temp = {'Distance': [d, d, d, d, d], 'Delay': [1, 2, 3, 4, 5]}
x1 = pd.DataFrame(data=temp)
y_pred = learning.knn_predict(knn_model, x1)
print("Prediction successful!")
print(y_pred)
idx = np.where(y_pred == np.amax(y_pred))
print(idx[0])
idx = np.amax(idx)
print(idx)
delay = idx + 1
print('Delay to be applied: ', delay)
####### Execute hit ########
ser = serial.Serial()
ser.baudrate = 9600
ser.port = 'com5'
ser.open()
i = bytearray([delay])
#send value here
ser.write(i)
print("sent")
ser.close()
'chopper': set_chopper_params,
'pong': set_pong_params
}
#core user interface dialouge
if len(sys.argv) == 3 and sys.argv[1] == 'stress':
mutation_count = int(sys.argv[2])
set_stress_test_params()
b = brain.Brain(1)
b.mutation_stress_test(mutation_count)
elif len(sys.argv) == 3 and sys.argv[1] == 'train':
param_setup = environment_settings[sys.argv[2]]
param_setup()
result = learning.learn(environments[sys.argv[2]])
elif len(sys.argv) == 4 and sys.argv[1] == 'analyze':
load_brain = brain.load_brain_from_file(sys.argv[3])
param_setup = environment_settings[sys.argv[2]]
param_setup()
learning.visualize_performance(load_brain, environments[sys.argv[2]])
elif len(sys.argv) == 4 and sys.argv[1] == 'improve':
load_brain = brain.load_brain_from_file(sys.argv[3])
param_setup = environment_settings[sys.argv[2]]
param_setup()
result = learning.learn_from_existing(load_brain,
environments[sys.argv[2]])
else:
print('INVALID USAGE')
def learning(self):
cores_path = str(self.text_cores.text())
target_path = str(self.text_target.text())
learning.learn(cores_path, target_path)
def learn(args):
"""
予測用モデルから予測値を取得
"""
learning.learn(args.start_date, args.end_date, args.var_lag)
ax1.imshow(np.reshape(train_set[NN, :], (Nr, Nc)),
cmap=plt.get_cmap('gray'),
aspect='auto'), ax1.set_title("Original Document " + str(Nr) + "x" +
str(Nc))
ax2.imshow(np.reshape(trfeat_set[NN, :rN**2], (rN, rN)),
cmap=plt.get_cmap('gray'),
aspect='auto'), ax2.set_title("Perceptual Hash " + str(rN) + "x" +
str(int(rN / 2)))
plt.colorbar()
plt.show()
#%% Learning on training set
import learning
# Only in case of phash or ahash or whash0;des
clasf = learning.learn(trfeat_set, y_train, method=c_method)
#%% Prediction on test set
# create_validation: Get batch of Nb images per class, full resolution
test_set, y_test, Nr, Nc = preproc.create_sample(dataset,
nitems,
label='test',
Nbatch=n_val)
tefeat_set = features.feat_extr(test_set,
Nr,
Nc,
feat_type=f_method,
hash_size=hash_s)
del test_set
def process(tag_name, tc, feature_func, save_flag=False, try_times=2000):
""" 学習とその検証を繰り返して、結果をファイルに保存する
"""
create_dir([tag_name])
with open(tag_name + "/av_verify_report" + tag_name + ".csv", "w") as fw:
for i in range(try_times):
print("--try count: {0}/{1}--".format(i, try_times))
# 教師データを作成
verify_data, teacher_dates, teacher_features, teacher_flags = None, None, None, None
if save_flag:
_save_name_teacher = tag_name + "/av_teaching_data" + tag_name + "_{0:05d}".format(i) + ".csv" # tc.save_teacher(ファイル名)で教師と検証データを保存するときに使う
_save_name_verify = tag_name + "/av_verify_data" + tag_name + "_{0:05d}".format(i) + ".csv"
verify_data, teacher_dates, teacher_features, teacher_flags = tc.save_teacher(save_name_teacher=_save_name_teacher, save_name_verify=_save_name_verify) # 毎回、呼び出すたびにデータセットの内容が変わる
else:
verify_data, teacher_dates, teacher_features, teacher_flags = tc.create_dataset() # 毎回、呼び出すたびにデータセットの内容が変わる
features_dict = tc.get_all_features()
training_data = (teacher_features, teacher_flags)
dates = sorted(verify_data.keys()) # 学習に使っていない日付
# 学習
clf = learning.learn(training_data, tag_name + "/av_entry_temp{0}_{1:05d}.pickle".format(tag_name, i))
result = predict.predict2(clf, dates, features_dict) # 学習に使っていないデータで検証
# 必要なら個別の結果も保存
if save_flag:
with open(tag_name + "/av_verify_data" + tag_name + "_{0:05d}".format(i) + "_result.csv", "w") as fw_result:
for date in dates:
if date in result:
try: # このtryはいらないんじゃないかな・・・
fw_result.write(str(date))
fw_result.write(",")
fw_result.write(str(verify_data[date]))
fw_result.write(",")
fw_result.write(str(result[date]))
fw_result.write("\n")
except:
pass
# 結果の集計
scale = 10
zero = [0.000001] * scale # sum()して分母に入れようとしたら、0の時にエラーが出るので0.000001とした
one = [0.000001] * scale
for date in dates:
if date in result:
try: # このtryはいらないんじゃないかな・・・
c = verify_data[date]
val = c - result[date]
if int(c) == 0:
zero[abs(int(val * scale))] += 1
elif int(c) == 1:
one[abs(int(val * scale))] += 1
#print(val)
except:
pass
# 最終結果の一覧ファイルへの保存
zero = [str(x / sum(zero)) for x in zero] # 正規化
one = [str(x / sum(one)) for x in one]
fw.write("{0},".format(i))
fw.write(",".join(zero))
fw.write(",,") # Excelで閲覧した時に分離させる
fw.write(",".join(one))
fw.write("\n")
def main(argus):
model = build_model()
kfold = argus.kfold
data_path = 'VeReMi-Dataset/' + argus.at + '_' + argus.t
learning.learn(model, data_path)
def relearn(self):
learning.learn()
raise cherrypy.HTTPRedirect("/stats")